13.3 Water Meters: Types, Accuracy & Meter Management
Key Takeaways
- Water Meters is a named sub-topic in both the treatment and the distribution Expected Range of Knowledge documents.
- Displacement meters measure small residential flows accurately, turbine meters handle high sustained flows, and compound meters combine both for variable demand.
- Meters almost always fail by under-registering as they wear, so an aging meter population produces apparent water loss and lost revenue.
- AWWA accuracy limits for new displacement meters are generally 98.5 to 101.5 percent at normal and maximum flows and at least 95 percent at minimum flow.
- Magnetic and ultrasonic meters have no moving parts and hold accuracy over wide flow ranges, but a magnetic meter requires a conductive fluid and full pipe.
Why Meters Are on Both Exams
Water Meters appears in the treatment Operation/Maintenance category (20 questions at T1 and T2) and in the distribution Equipment category (20 to 25 questions across the grades). The treatment side is about process flow measurement; the distribution side is about customer metering and water loss. Both come down to the same question: do you believe the number, and how would you prove it?
Customer Meter Technologies
| Type | Principle | Best for | Weakness |
|---|---|---|---|
| Positive displacement (nutating disc, oscillating piston) | Each revolution passes a fixed volume | Residential and small commercial; excellent low-flow accuracy | Limited capacity; wears and under-registers; damaged by grit |
| Single-jet / multi-jet | Jets impinge on an impeller | Small to medium services | Sensitive to debris |
| Turbine | Axial rotor in the stream | High, steady flows - industrial, irrigation, master meters | Poor low-flow accuracy; misses small leaks entirely |
| Compound | A displacement element for low flow plus a turbine for high flow, with a changeover valve | Highly variable demand - hotels, hospitals, schools | Complex; the changeover valve is a failure point |
| Fire service meter | Low head loss at very high flow, with a bypass metering element | Fire lines with incidental domestic use | Must pass fire flow with minimal loss |
| Electromagnetic (mag) | Faraday induction; voltage proportional to velocity | Plant process flow, master meters, dirty water and wastewater | Requires a conductive fluid and a full pipe; needs power |
| Ultrasonic (transit time) | Time difference of sound with and against flow | Wide range, no moving parts; increasingly used for residential | Sensitive to air and heavy solids |
Accuracy and Testing
Meters are tested at three flows, and the standards differ for new and in-service meters. For new displacement meters, AWWA accuracy limits are generally:
| Test flow | Acceptable accuracy (new displacement meter) |
|---|---|
| Minimum (low) flow | 95 to 101 percent |
| Intermediate (normal) flow | 98.5 to 101.5 percent |
| Maximum flow | 98.5 to 101.5 percent |
Worked example. A 5/8-inch meter is tested by passing a measured 10.00 cubic feet through it; the meter registers 9.62 cubic feet. Accuracy = 9.62 / 10.00 x 100 = 96.2 percent - acceptable at the low-flow test point but failing at the intermediate or maximum test point.
[!IMPORTANT] Meters wear toward under-registration, not over-registration. A worn nutating disc slips; a worn turbine bearing drags. That means an aging meter population systematically under-bills customers and inflates apparent water loss. When a system's non-revenue water climbs and leak detection finds nothing, the meters are the first suspect. The rare over-registration causes are air passing through the meter and, on some meter types, reverse flow being counted as forward flow.
Testing Intervals
| Meter size | Typical test/replacement interval |
|---|---|
| 5/8-inch and 3/4-inch residential | Test or replace on a rotating schedule, commonly 10 to 20 years or at a registered-volume threshold |
| 1-inch to 2-inch | 6 to 10 years |
| 3-inch and larger | Annually to every 2 years - large meters carry most of the revenue risk |
| Master and source meters | Annually - these anchor the entire water balance |
Large meters justify frequent testing because a 1 percent error on a 4-inch industrial meter is worth far more than a 10 percent error on a residential meter.
Installation Requirements
- Straight pipe runs. Turbine, compound, mag, and ultrasonic meters need upstream and downstream straight lengths (commonly 5 to 10 pipe diameters upstream and 2 to 5 downstream) so the velocity profile is developed. Installing a turbine directly downstream of an elbow guarantees error.
- Full pipe. Mag and ultrasonic meters must run full; a partially filled pipe reads nonsense. Install in a low point or on a riser.
- Correct orientation and flow direction arrow.
- Strainer ahead of turbine and compound meters.
- Bypass on large meters so the customer is not shut off during testing.
- Backflow protection appropriate to the hazard - typically a dual check on residential services, an assembly on higher-hazard services.
- Accessible box or vault, drained and safe to enter, with the register readable.
Reading Systems
| System | Description |
|---|---|
| Manual read | Meter reader opens the box and reads the register |
| Touch read | A probe contacts a pad wired to the register |
| AMR (automatic meter reading) | Drive-by or walk-by radio collection |
| AMI (advanced metering infrastructure) | Fixed network with hourly or 15-minute interval data, two-way communication |
AMI changes operations, not just billing. Interval data lets a utility detect continuous flow at a service (a running toilet or a service leak) and notify the customer within days rather than at the next quarterly bill; it supports district metered area leak detection; and it provides the demand data that 22 CCR 64554 wants for computing maximum day and peak hour demand from real records rather than a 2.25 default factor.
Water Loss Accounting
The AWWA water audit splits losses into:
- Apparent losses - meter under-registration, data handling errors, unauthorized consumption. These are billed-revenue losses at retail value.
- Real losses - leakage on mains, services, and storage overflows. These are production-cost losses.
California requires urban retail water suppliers to submit an annual validated water loss audit to the Department of Water Resources, so meter accuracy has become a state reporting matter, not just a billing preference. A meter replacement program is a water loss control program.
Plant Flow Measurement
The treatment-side blueprint entry covers process flow devices, which share the same principles:
| Device | Principle | Notes |
|---|---|---|
| Venturi | Differential pressure across a converging-diverging throat | Low permanent head loss; very durable |
| Orifice plate | Differential pressure across a plate | Cheap; high permanent head loss |
| Magnetic flow meter | Faraday induction | No obstruction, no head loss; needs conductivity and a full pipe |
| Ultrasonic | Transit time or Doppler | Clamp-on versions need no pipe penetration |
| Propeller / turbine | Rotor speed | Common on wells; needs straight run |
| Parshall flume | Open channel, critical depth at a throat | Self-cleaning, standard in wastewater |
| Weir (rectangular, V-notch) | Head over a crest | V-notch is best for low flows; weirs accumulate solids |
[!TIP] For open-channel devices the flow is computed from head above the crest or floor, so a level error is a flow error. Verify the zero, keep the approach channel clean, and confirm that the flume or weir is not submerged by downstream backwater, which invalidates the rating entirely.
A test passes a measured 10.00 cubic feet of water through a residential meter, which registers 9.55 cubic feet. What is the meter accuracy and what does it imply?
A hotel has demand that swings from a trickle overnight to very high flows at peak times. Which meter type is most appropriate?
A newly installed electromagnetic flow meter on a plant process line gives erratic readings. Which installation condition should be checked first?